
About this episode
Rohan Kodialam, cofounder and CEO of Sphinx, is building AI agents that treat data as its own language—one most models and humans still fail to understand. In this episode, he unpacks why data science has lagged behind software engineering, how AI can finally close the gap between business questions and answers, and what happens when small teams gain the analytical power of a thousand person quant desk.
What You'll Learn
• How AI models that actually see data can unlock insights traditional transformers miss
• Why enterprises must rethink dashboards and embrace real time ad hoc analysis
• Where AI truly saves the most time across the data lifecycle and why modeling is not the hardest part
• How decoupling statistics from business context gives teams freedom to focus on strategy and creativity
• Why success in data science now means reclaiming human creativity while automating repetitive work
Timestamped Highlights
[01:44] Why data is fundamentally different from text and code and why most AI models struggle with it
[06:39] The cultural problem with ad hoc being a dirty word in enterprises and why that mindset is changing
[11:09] Where AI tools actually fit into the data science workflow
[17:09] How to measure success when using an AI data scientist
[21:04] What happens when a small team gains the data firepower of a hedge fund quant operation
[24:37] Why bad data science is worse than none and why quality matters more than hype
A Thought That Stuck With Us
“We are cutting the time to completion by 20x, 40x, even 50x and that remaining human review is not a bottleneck. It is the feature that keeps AI accountable.”
Worth Following
Connect with Rohan Kodialam on X (@KodialamRo) or LinkedIn and learn more about Sphinx AI and how they are transforming enterprise data science.
If This Resonated
Share this with someone in the data world who is tired of waiting weeks for insights that should take minutes. Follow The Tech Trek for more conversations about how people and technology create lasting impact.
Get every episode summarized
Each time The Tech Trek publishes, we email you a written briefing from the transcript — the topics, who appeared, and any specific claims, with the ad reads skipped.
Email me new episodesFree for 3 shows. No card needed.
Hosts & guests
No transcript yet
This episode has not been transcribed. Request it and it moves to the front of the queue.
More episodes
More from The Tech Trek

How AI Is Changing Engineering Workflows and Software Teams
The Tech Trek

How AI Agents Are Changing Who Can Build Software
The Tech Trek

How to Build a Data Team Ready for AI
The Tech Trek

How AI Startups Can Challenge the Enterprise Services Model
The Tech Trek